Application of Psychometric Methods in Dimensional Analysis and Integration of Assessment Tools in Early Diagnosis for Autism Spectrum Disorder

Background: Early diagnosis of autism spectrum disorder (ASD) is crucial for timely intervention, and requires reliable and valid screening and diagnostic tools. The Toddler Module of the Autism Diagnostic Observation Schedule, Second Edition (ADOS-2), is widely used but its factor structure and applicability in the clinical practice need further investigation. Objective: This study aimed to explore the underlying dimensional structure of the ADOS-2 Toddler Module in Italian context, and to examine the association between the identified ASD symptom dimensions and cognitive development, as measured by the Griffiths Mental Development Scales (GMDS). Methods: The study was conducted in two phases: in Phase 1, Exploratory Graph Analysis (EGA) was used to identify latent dimensions of the ADOS-2 Toddler Module items in a sample of 91 Italian children aged 12–30 months at risk for ASD. In Phase 2, structural equation modeling (SEM) was performed on a subsample of 60 children who completed the GMDS to examine associations between ASD symptom dimensions and cognitive development. Results: EGA revealed a stable three-factor structure underlying the Toddler Module items. The SEM analysis demonstrated a good model fit and significant associations between ASD symptom dimensions and the Personal-Social subscale of the GMDS. Conclusions: These findings provide preliminary evidence for a three-dimensional structure of ASD symptoms in very young children and suggest that cognitive development, particularly social-personal skills, is related to early ASD symptomatology. The results have potential implications for refining early diagnostic assessments and guiding clinical practice.